1. Field of the Invention
This invention relates to an inductive device, and more particularly, to a high efficiency, low core loss inductive device having a core comprising one or more bulk amorphous metal magnetic components.
2. Description of the Prior Art
Inductive devices are essential components of a wide variety of modern electrical and electronic equipment, most commonly including transformers and inductors. Most of these devices employ a core comprising a soft ferromagnetic material and one or more electrical windings that encircle the core. Inductors generally employ a single winding with two terminals, and serve as filters and energy storage devices. Transformers generally have two or more windings. They transform voltages from one level to at least one other desired level, and electrically isolate different portions of an overall electric circuit. Inductive devices are available in widely varying sizes with correspondingly varying power capacities. Different types of inductive devices are optimized for operation at frequencies over a very wide range, from DC to GHz. Virtually every known type of soft magnetic material finds application in the construction of inductive devices. Selection of a particular soft magnetic material depends on the combination of properties needed, the availability of the material in a form that lends itself to efficient manufacture, and the volume and cost required to serve a given market. In general, a desirable soft ferromagnetic core material has high saturation induction Bsat to minimize core size, and low coercivity Hc, high magnetic permeability μ, and low core loss to maximize efficiency.
Components such as motors and small to moderate size inductors and transformers for electrical and electronic devices often are constructed using laminations punched from various grades of magnetic steel supplied in sheets having thickness as low as 100 μm. The laminations are generally stacked and secured and subsequently wound with the requisite one or more electrical windings that typically comprise high conductivity copper or aluminum wire. These laminations are commonly employed in cores with a variety of known shapes.
Many of the shapes used for inductors and transformers are assembled from constituent components which have the general form of certain block letters, such as “C,” “U,” “E,” and “I”, by which the components are often identified. The assembled shape may further be denoted by the letters reflecting the constituent components; for example, an “E-I” shape would be made by assembling an “E” component with an “I” component. Other widely used assembled shapes include “E-E,” “C-I,” and “C-C.” Constituent components for prior art cores of these shapes have been constructed both of laminated sheets of conventional crystalline ferromagnetic metal and of machined bulk soft ferrite blocks.
Although many amorphous metals offer superior magnetic performance when compared to other common soft ferromagnetic materials, certain of their physical properties make conventional fabrication techniques difficult or impossible. Amorphous metal is typically supplied as a thin, continuous ribbon having a uniform ribbon width. However, amorphous metals are thinner and harder than virtually all conventional metallic soft magnetic alloys, so conventional stamping or punching of laminations causes excessive wear on fabrication tools and dies, leading to rapid failure. The resulting increase in the tooling and manufacturing costs makes fabricating bulk amorphous metal magnetic components using such conventional techniques commercially impractical. The thinness of amorphous metals also translates into an increased number of laminations needed to form a component with a given cross-section and thickness, further increasing the total cost of an amorphous metal magnetic component. Machining techniques used for shaping ferrite blocks are also not generally suited for processing amorphous metals.
The properties of amorphous metal are often optimized by an annealing treatment. However, the annealing generally renders the amorphous metal very brittle, further complicating conventional manufacturing processes. As a result of the aforementioned difficulties, techniques that are widely and readily used to form shaped laminations of silicon steel and other similar metallic sheet-form FeNi- and FeCo-based crystalline materials, have not been found suitable for manufacturing amorphous metal devices and components. Amorphous metals thus have not been accepted in the marketplace for many devices; this is so, notwithstanding the great potential for improvements in size, weight, and energy efficiency that in principle would be realized from the use of a high induction, low loss material.
For electronic applications such as saturable reactors and some chokes, amorphous metal has been employed in the form of spirally wound, round toroidal cores. Devices in this form are available commercially with diameters typically ranging from a few millimeters to a few centimeters and are commonly used in switch-mode power supplies providing up to several hundred volt-amperes (VA). This core configuration affords a completely closed magnetic circuit, with negligible demagnetizing factor. However, in order to achieve a desired energy storage capability, many inductors require a magnetic circuit that includes a discrete air gap. The presence of the gap results in a non-negligible demagnetizing factor and an associated shape anisotropy that are manifested in a sheared magnetization (B-H) loop. The shape anisotropy may be much higher than the possible induced magnetic anisotropy, increasing the energy storage capacity proportionately. Toroidal cores with discrete air gaps and conventional material have been proposed for such energy storage applications.
However, the stresses inherent in a strip-wound toroidal core give rise to certain problems. The winding inherently places the outside surface of the strip in tension and the inside in compression. Additional stress is contributed by the linear tension needed to insure smooth winding. As a consequence of magnetostriction, a wound toroid typically exhibits magnetic properties that are inferior to those of the same strip measured in a flat strip configuration. Annealing in general is able to relieve only a portion of the stress, so only a part of the degradation is eliminated. In addition, gapping a wound toroid frequently causes additional problems. Any residual hoop stress in the wound structure is at least partially removed on gapping. In practice the net hoop stress is not predictable and may be either compressive or tensile. Therefore the actual gap tends to close or open in the respective cases by an unpredictable amount as required to establish a new stress equilibrium. Therefore, the final gap is generally different from the intended gap, absent corrective measures. Since the magnetic reluctance of the core is determined largely by the gap, the magnetic properties of finished cores are often difficult to reproduce on a consistent basis in the course of high-volume production.
Furthermore, designers frequently seek flexibility not afforded by a limited selection of standard gapped toroidal core structures. For these applications, it is desirable for a user to be able to adjust the gap so as to select a desired degree of shearing and energy storage. In addition, the equipment needed to apply windings to a toroidal core is more complicated, expensive, and difficult to operate than comparable winding equipment for laminated cores. Oftentimes a core of toroidal geometry cannot be used in a high current application, because the heavy gage wire dictated by the rated current cannot be bent to the extent needed in the winding of a toroid. In addition, toroidal designs have only a single magnetic circuit. As a result, they are generally best suited for single phase applications. Other configurations more amenable to easy manufacture and application, while still affording attractive magnetic properties and efficiency, especially for polyphase (including three phase) requirements, are thus sought.
Amorphous metals have also been used in transformers for much higher power devices, such as distribution transformers for the electric power grid that have nameplate ratings of 10 kVA to 1 MVA or more. The cores for these transformers are often formed in a step-lap wound, generally rectangular configuration. In one common construction method, the rectangular core is first formed and annealed. The core is then unlaced to allow pre-formed windings to be slipped over the long legs of the core. Following the incorporation of the pre-formed windings, the layers are relaced and secured. A typical process for constructing a distribution transformer in this manner is set forth in U.S. Pat. No. 4,734,975 to Ballard et al. Such a process understandably entails significant manual labor and manipulation steps involving brittle annealed amorphous metal ribbons. These steps are especially tedious and difficult to accomplish with cores smaller than 10 kVA. Furthermore, in this configuration, the cores are not readily susceptible to controllable introduction of an air gap, which is needed for many inductor applications.
Another difficulty associated with the use of ferromagnetic amorphous metals arises from the phenomenon of magnetostriction. Certain magnetic properties of any magnetostrictive material change in response to imposed mechanical stress. For example, the magnetic permeability of a component containing amorphous materials typically is reduced, and its core losses are increased, when the component is subjected to stress. The degradation of soft magnetic properties of the amorphous metal device due to the magnetostriction phenomenon may be caused by stresses resulting from any combination of sources, including deformation during core fabrication, mechanical stresses resulting from mechanical clamping or otherwise fixing the amorphous metal in place and internal stresses caused by the thermal expansion and/or expansion due to magnetic saturation of the amorphous metal material. As an amorphous metal magnetic device is stressed, the efficiency at which it directs or focuses magnetic flux is reduced, resulting in higher magnetic losses, reduced efficiency, increased heat production, and reduced power. The extent of this degradation is oftentimes considerable. It depends upon the particular amorphous metal material and the actual intensity of the stresses, as indicated by U.S. Pat. No. 5,731,649.
Amorphous metals have far lower anisotropy energies than many other conventional soft magnetic materials, including common electrical steels. Stress levels that would not have a deleterious effect on the magnetic properties of these conventional metals have a severe impact on magnetic properties such as permeability and core loss, which are important for inductive components. For example, the '649 patent teaches that forming amorphous metal cores by rolling amorphous metal into a coil, with lamination using an epoxy, detrimentally restricts the thermal and magnetic saturation expansion of the coil of material. High internal stresses and magnetostriction are thereby produced, which reduce the efficiency of a motor or generator incorporating such a core. In order to avoid stress-induced degradation of magnetic properties, the '649 patent discloses a magnetic component comprising a plurality of stacked or coiled sections of amorphous metal carefully mounted or contained in a dielectric enclosure without the use of adhesive bonding.
A significant trend in recent technology has been the design of power supplies, converters, and related circuits using switch-mode circuit topologies. The increased capabilities of available power semiconductor switching devices have allowed switch-mode devices to operate at increasingly high frequencies. Many devices that formerly were designed with linear regulation and operation at line frequencies (generally 50-60 Hz on the power grid or 400 Hz in military applications) are now based on switch-mode regulation at frequencies that are often 5-200 kHz, and sometimes as much as 1 MHz. A principal driving force for the increase in frequency is the concomitant reduction in the size of the required magnetic components, such as transformers and inductors. However, the increase in frequency also markedly increases the magnetic losses of these components. Thus there exists a significant need to lower these losses.
The limitations of magnetic components made using existing materials entail substantial and undesirable design compromises. In many applications, the core losses of the common electrical steels are prohibitive. In such cases a designer may be forced to use a permalloy alloy or a ferrite as an alternative. However, the attendant reduction in saturation induction (e.g. 0.6-0.9 T or less for various permalloy alloys and 0.3-0.4 T for ferrites, versus 1.8-2.0 T for ordinary electrical steels) necessitates an increase in the size of the resulting magnetic components. Furthermore, the desirable soft magnetic properties of the permalloys are adversely and irreversibly affected by plastic deformation which can occur at relatively low stress levels. Such stresses may occur either during manufacture or operation of the permalloy component. While soft ferrites often have attractively low losses, their low induction values result in impractically large devices for many applications wherein space is an important consideration. Moreover, the increased size of the core undesirably necessitates a longer electrical winding, so ohmic losses increase.
Notwithstanding the advances represented by the above disclosures, there remains a need in the art for improved inductive devices that exhibit a combination of excellent magnetic and physical properties needed for current requirements. Construction methods are also sought that use amorphous metal efficiently and can be implemented for high volume production of devices of various types.